Where is that value premium?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
KJVanguard
Posts: 61
Joined: Fri Oct 05, 2018 3:39 am

Where is that value premium?

Post by KJVanguard » Fri Oct 05, 2018 4:18 am

Vanguard has offered a Small Growth & Small Value Index since May 1998, so we now have just over 20 years of real world performance to look at. Theory is lovely, but real world results are the only thing that can fill your wallet.

Over the last 20 years (or so):

SG = 9.26%
SV = 8.93%

Well, this isn't the way I'm told it's supposed to be, so what is going on? Why do we have a 0.33% annual premium for Small Growth, an investment that would also have cost you less in taxes due to its lower dividends (and thus higher cap gains that get preferred tax treatment).

Let's consider the possibilities, or at least all that I can think of and you can tell me which one(s) you think are correct:

1. There is no value premium and that's exactly what we are seeing here.

2. There is a value premium, but one can't capture it with a Vanguard fund as they are not "valuey" enough. Certainly possible since Fama French seem to be talking about DEEP VALUE, something you won't find at Vanguard or substantially anywhere else for that matter. I guess one could pay Rick Ferri (being the lowest priced advisor) to get access to DEEP VALUE DFA funds, though that would cost money (even if modest) and that would eat into your value premium. I don't think Rick would even advise use of his firm ONLY to get into DFA funds. Obviously, Rick is much more than a mere DFA gate-keeper so he could add value for the average investor, though if you are reading this odds are you are well beyond average.

3. There is a value premium, but it just failed to show up over a span of "only" 20 years. Certainly possible, but knowing I'm going to win in the "long run" is of little comfort to us mere mortals who will all be dead in the long run. Hey, I mean the equity premium has failed to show up in Japan in the last three decades, but we all still believe in it. Wonder if investors in Japan believe in it, with many not being old enough to remember it?

There seems to be compelling academic evidence of a historical value premium across markets around the world, so I would tend to believe it exists. Just one problem: how do I get my hands on that value premium? One can look for deep value investors, though most of these active deep value investors vividly demonstrate how it's damn hard to just to keep up with the S&P 500, much less deliver a premium of any sort.

Should we just give up on this elusive quest to capture the value premium, seeing how there seems to be no practical way for an investor to actually get it? This is similar to how 10th decile stocks might beat the market, but any attempt to actually buy these nano-cap stocks proves an impossible task. Even micro caps funds tend to stray into small cap territory, away from the magical 10th decile.

Thank you for reading. Please share your thoughts with me.

livesoft
Posts: 62717
Joined: Thu Mar 01, 2007 8:00 pm

Re: Where is that value premium?

Post by livesoft » Fri Oct 05, 2018 5:53 am

I think there are no premiums for folks who buy once and hold forever or hold for 20 years.
Wiki This signature message sponsored by sscritic: Learn to fish.

wolf359
Posts: 1390
Joined: Sun Mar 15, 2015 8:47 am

Re: Where is that value premium?

Post by wolf359 » Fri Oct 05, 2018 6:15 am

From Investopedia:
Risk:
Risk takes on many forms but is broadly categorized as the chance an outcome or investment's actual return will differ from the expected outcome or return. Risk includes the possibility of losing some or all of the original investment.

Read more: Risk https://www.investopedia.com/terms/r/ri ... z5T3PHIbUc
Follow us: Investopedia on Facebook
Small cap value has a higher expected return, but it is an expectation, not a written guarantee. There IS the possibility that the premium might not materialize. If it were guaranteed, everybody would invest in it.

Also, I believe the premium is being measured relative to beta, not SG versus SV. SG is more about lottery stocks, and I'd expect them to be peaking at the top of a long bull market (such as where we are today.)

Jags4186
Posts: 2478
Joined: Wed Jun 18, 2014 7:12 pm

Re: Where is that value premium?

Post by Jags4186 » Fri Oct 05, 2018 6:16 am

Since June 1998:

Vanguard Small Cap Index Fund: 8.80% CAGR
Vanguard Small Cap Value Index Fund: 9.06% CAGR
Vanguard Small Cap Growth Index Fund: 9.45% CAGR
Vanguard Value Index Fund: 6.98% CAGR (since August 1998)
Vanguard Growth Index Fund: 7.12% CAGR
Vanguard 500 Index Fund: 6.85% CAGR

There was a value premium and there was a small premium. Growth just happened to perform quite well during this period. Also keep in mind Vanguards SCV and SCG funds are not very small and not very tilted.

DFA Small Cap Value Fund since June 1998: 9.87% CAGR

User avatar
JoMoney
Posts: 5917
Joined: Tue Jul 23, 2013 5:31 am

Re: Where is that value premium?

Post by JoMoney » Fri Oct 05, 2018 6:51 am

KJVanguard wrote:
Fri Oct 05, 2018 4:18 am
...
Well, this isn't the way I'm told it's supposed to be, so what is going on? ...
Maybe you're listening to the wrong people.
John Bogle in 1996 annual report wrote: https://www.sec.gov/Archives/edgar/data ... 000339.txt

... On this note, you may recall that I mentioned to you one year ago that the various market segments seem to enjoy cycles of superiority in their (unpredictable) turns. But, over longer periods, their respective returns tend to converge. Surely, both the interim and longer-run patterns are demonstrated in the charts above. As a result, I believe it is fair to conclude that, for most investors, the 500 Portfolio and the Total Stock Market Portfolio should represent the preferred investment approaches. (The Standard & Poor's 500 Index comprises about 70% of the total market value of the entire stock market.)
For investors with particular objectives, our other Portfolios--Growth, Value, Extended Market, and SmallCap--assure broad diversification in discrete market segments. To be sure, investors must stick to their objectives with consistency and firmness, ignoring the siren songs that suggest that moving money back and forth from one segment to another will result in a sustainable performance advantage. Such "market timing" is all too likely to be self-defeating. Whichever of the Trust's six Portfolios you have selected, the long-term, steady-as-she-goes approach is virtually certain to be the most productive strategy. ...
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

User avatar
JoMoney
Posts: 5917
Joined: Tue Jul 23, 2013 5:31 am

Re: Where is that value premium?

Post by JoMoney » Fri Oct 05, 2018 7:00 am

Looking at the Fama-French data, there hasn't been a persistent "Value Premium" in large stocks in over 40 years
Image
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

beardsworth
Posts: 2135
Joined: Fri Jun 15, 2007 4:02 pm

Re: Where is that value premium?

Post by beardsworth » Fri Oct 05, 2018 7:02 am

The value premium is like Bigfoot or the Yeti. It goes long periods of time without being seen. :)

rkhusky
Posts: 5551
Joined: Thu Aug 18, 2011 8:09 pm

Re: Where is that value premium?

Post by rkhusky » Fri Oct 05, 2018 7:05 am

You can't just look at one time period, with a single start and a single end point and draw overall conclusions. For example, there are periods where bonds out-performed stocks. If you only looked at those periods (or happened to be doing your analysis at the end of one of those periods), you might conclude that you should prefer bonds over stocks for higher returns.

On the other hand, it does seem like factors get proposed, everyone jumps in, the factor premium disappears, a different factor gets proposed, ...

User avatar
vineviz
Posts: 1745
Joined: Tue May 15, 2018 1:55 pm

Re: Where is that value premium?

Post by vineviz » Fri Oct 05, 2018 11:36 am

rkhusky wrote:
Fri Oct 05, 2018 7:05 am
You can't just look at one time period, with a single start and a single end point and draw overall conclusions. For example, there are periods where bonds out-performed stocks. If you only looked at those periods (or happened to be doing your analysis at the end of one of those periods), you might conclude that you should prefer bonds over stocks for higher returns.
I agree with you here.

We are currently 10+ years into a period in which value stocks have generally underperformed the broad market. As a result, I think it takes a lot of discipline to refrain from pronouncing the value "premium" as permanently dead.

For instance, using Vanguard Value Index Fund (VIVAX) as proxy the cumulative value premium since that fund's inception in 1992 (measure as VIVAX return minus VFINX return) is indeed moderately negative at present.

Image

But the premium has been MUCH more negative before (1993 to 2000) but turned significantly positive again during the subsequent 2000 to 2007 period. We have no way of knowing if or when the premium will turn positive again, but we do have cavalcade of evidence to support an expectation that it will at some point.

In any case, if someone tried to tell us that the past ten years are evidence that "growth" stocks will permanently outperform the broad market I think that most Bogleheads would immediately recognize the fallacy there. Yet those same Bogleheads seem willing to rely on the same evidence to pronounce the value (or small size) effects to be permanently erased.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

afan
Posts: 3844
Joined: Sun Jul 25, 2010 4:01 pm

Re: Where is that value premium?

Post by afan » Fri Oct 05, 2018 11:37 am

Fama and French do not claim that value produces increased risk adjusted returns. They say that the higher expected returns are compensation for higher expected risk.

I don't get the fascination with getting market risk adjusted returns through value, rather than total market.
We don't know how to beat the market on a risk-adjusted basis, and we don't know anyone that does know either | --Swedroe | We assume that markets are efficient, that prices are right | --Fama

User avatar
vineviz
Posts: 1745
Joined: Tue May 15, 2018 1:55 pm

Re: Where is that value premium?

Post by vineviz » Fri Oct 05, 2018 11:44 am

afan wrote:
Fri Oct 05, 2018 11:37 am
I don't get the fascination with getting market risk adjusted returns through value, rather than total market.
I'm not sure I'd describe it as a "fascination" as much as I'd describe it as an acknowledgment that all investors don't have the same desire or ability to accept risk, in whatever form that risk might appear.

If my risk tolerance is higher than yours, for instance, it's completely rational for me to take on more risk than you even if (especially if) we are both getting the same risk-adjusted return.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
zonto
Posts: 119
Joined: Tue Aug 24, 2010 3:45 pm
Location: Boston, MA

Re: Where is that value premium?

Post by zonto » Fri Oct 05, 2018 4:41 pm

Thought I'd share this article from a couple years ago: Is the Value Premium Disappearing?. It also analyzes the value premium in developed and emerging international markets. Also this thread: viewtopic.php?t=191001, which analyzes the relationship between the value premium and inflation rates.
“Diversification is about accepting good enough while missing out on great but avoiding terrible.” - Ben Carlson

GammaPoint
Posts: 2608
Joined: Sun Aug 02, 2009 10:25 am
Location: Washington

Re: Where is that value premium?

Post by GammaPoint » Fri Oct 05, 2018 4:57 pm

livesoft wrote:
Fri Oct 05, 2018 5:53 am
I think there are no premiums for folks who buy once and hold forever or hold for 20 years.
I know you've mentioned this elsewhere before, but I've forgotten. How do you expect to harvest the value premium that you have in your portfolio (IIRC, you do have at least SV in your portfolio)? Or do you not expect a premium and just use it for the diversification benefits?

Elysium
Posts: 1261
Joined: Mon Apr 02, 2007 6:22 pm

Re: Where is that value premium?

Post by Elysium » Fri Oct 05, 2018 5:01 pm

I have a simple rule. Never complain when an asset I hold has returned 9% over 20 years!

Even if that means other assets may have gained more. I have held a SV fund (combination of Vanguard and DFA overlapping years) for close to 15 years and it (they) has made over 9% I think. I also hold large growth funds, Primecap, Capital Opp, etc and they have made more than 15% annualized I think.

Should I sell my SV fund and buy the large growth funds now? Nah, I think diversification is good, and I am getting enough. That's what matters.

livesoft
Posts: 62717
Joined: Thu Mar 01, 2007 8:00 pm

Re: Where is that value premium?

Post by livesoft » Fri Oct 05, 2018 5:16 pm

GammaPoint wrote:
Fri Oct 05, 2018 4:57 pm
livesoft wrote:
Fri Oct 05, 2018 5:53 am
I think there are no premiums for folks who buy once and hold forever or hold for 20 years.
I know you've mentioned this elsewhere before, but I've forgotten. How do you expect to harvest the value premium that you have in your portfolio (IIRC, you do have at least SV in your portfolio)? Or do you not expect a premium and just use it for the diversification benefits?
I expect to harvest the value premium when value index funds underperform the total market funds, so that I can buy more of the value index fund shares, then wait until they catch up and outperform the total market funds, then rebalance a bit out of the value index funds. Rinse and repeat ad nauseam.
Wiki This signature message sponsored by sscritic: Learn to fish.

GammaPoint
Posts: 2608
Joined: Sun Aug 02, 2009 10:25 am
Location: Washington

Re: Where is that value premium?

Post by GammaPoint » Fri Oct 05, 2018 7:30 pm

livesoft wrote:
Fri Oct 05, 2018 5:16 pm
I expect to harvest the value premium when value index funds underperform the total market funds, so that I can buy more of the value index fund shares, then wait until they catch up and outperform the total market funds, then rebalance a bit out of the value index funds. Rinse and repeat ad nauseam.
Got it. Thanks.

Wakefield1
Posts: 839
Joined: Mon Nov 14, 2016 10:10 pm

Re: Where is that value premium?

Post by Wakefield1 » Fri Oct 05, 2018 8:19 pm

Say a fund at a point in time shows superior 10 year return. Then suppose it has a really bad quarter and underperforms its peers by several percent. I understand that the 10 year return figure will also decline relative to the other funds that did better that quarter. There goes the (value) (growth) (small) (large cap) or whatever premium?

afan
Posts: 3844
Joined: Sun Jul 25, 2010 4:01 pm

Re: Where is that value premium?

Post by afan » Fri Oct 05, 2018 8:20 pm

vineviz wrote:
Fri Oct 05, 2018 11:44 am


If my risk tolerance is higher than yours, for instance, it's completely rational for me to take on more risk than you even if (especially if) we are both getting the same risk-adjusted return.
Sure. But you can get that by increasing your stock allocation in VTI. No need to go to play with value funds.
We don't know how to beat the market on a risk-adjusted basis, and we don't know anyone that does know either | --Swedroe | We assume that markets are efficient, that prices are right | --Fama

User avatar
watchnerd
Posts: 1428
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Where is that value premium?

Post by watchnerd » Fri Oct 05, 2018 8:56 pm

afan wrote:
Fri Oct 05, 2018 8:20 pm
vineviz wrote:
Fri Oct 05, 2018 11:44 am


If my risk tolerance is higher than yours, for instance, it's completely rational for me to take on more risk than you even if (especially if) we are both getting the same risk-adjusted return.
Sure. But you can get that by increasing your stock allocation in VTI. No need to go to play with value funds.
Yeah, I don't get it, either.

If I want more risk / more return, I increase my equities and decrease my bonds.
Tax Sheltered: 35% US Stock | 35% ex-US Stock | 30% TTM || Taxable: 35% US Stock | 35% ex-US Stock | 15% TTM | 15% Munis

MotoTrojan
Posts: 2294
Joined: Wed Feb 01, 2017 8:39 pm

Re: Where is that value premium?

Post by MotoTrojan » Fri Oct 05, 2018 9:49 pm

watchnerd wrote:
Fri Oct 05, 2018 8:56 pm
afan wrote:
Fri Oct 05, 2018 8:20 pm
vineviz wrote:
Fri Oct 05, 2018 11:44 am


If my risk tolerance is higher than yours, for instance, it's completely rational for me to take on more risk than you even if (especially if) we are both getting the same risk-adjusted return.
Sure. But you can get that by increasing your stock allocation in VTI. No need to go to play with value funds.
Yeah, I don't get it, either.

If I want more risk / more return, I increase my equities and decrease my bonds.
Well I’m 100/0. Now what?

User avatar
watchnerd
Posts: 1428
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Where is that value premium?

Post by watchnerd » Fri Oct 05, 2018 11:00 pm

MotoTrojan wrote:
Fri Oct 05, 2018 9:49 pm
watchnerd wrote:
Fri Oct 05, 2018 8:56 pm
afan wrote:
Fri Oct 05, 2018 8:20 pm
vineviz wrote:
Fri Oct 05, 2018 11:44 am


If my risk tolerance is higher than yours, for instance, it's completely rational for me to take on more risk than you even if (especially if) we are both getting the same risk-adjusted return.
Sure. But you can get that by increasing your stock allocation in VTI. No need to go to play with value funds.
Yeah, I don't get it, either.

If I want more risk / more return, I increase my equities and decrease my bonds.
Well I’m 100/0. Now what?
Leverage.
Tax Sheltered: 35% US Stock | 35% ex-US Stock | 30% TTM || Taxable: 35% US Stock | 35% ex-US Stock | 15% TTM | 15% Munis

AlphaLess
Posts: 742
Joined: Fri Sep 29, 2017 11:38 pm
Location: Kentucky
Contact:

Re: Where is that value premium?

Post by AlphaLess » Fri Oct 05, 2018 11:18 pm

KJVanguard wrote:
Fri Oct 05, 2018 4:18 am
SG = 9.26%
SV = 8.93%

Well, this isn't the way I'm told it's supposed to be.
Who said it isn't the way it's supposed to be?
"You can get more with a kind word and a gun than with just a kind word." George Washington

MotoTrojan
Posts: 2294
Joined: Wed Feb 01, 2017 8:39 pm

Re: Where is that value premium?

Post by MotoTrojan » Sat Oct 06, 2018 1:49 am

watchnerd wrote:
Fri Oct 05, 2018 11:00 pm
MotoTrojan wrote:
Fri Oct 05, 2018 9:49 pm
watchnerd wrote:
Fri Oct 05, 2018 8:56 pm
afan wrote:
Fri Oct 05, 2018 8:20 pm
vineviz wrote:
Fri Oct 05, 2018 11:44 am


If my risk tolerance is higher than yours, for instance, it's completely rational for me to take on more risk than you even if (especially if) we are both getting the same risk-adjusted return.
Sure. But you can get that by increasing your stock allocation in VTI. No need to go to play with value funds.
Yeah, I don't get it, either.

If I want more risk / more return, I increase my equities and decrease my bonds.
Well I’m 100/0. Now what?
Leverage.
Why is that better than small-cap? Why wouldn’t I leverage a 90/10 or 60/40 instead of 100/0?

User avatar
watchnerd
Posts: 1428
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Where is that value premium?

Post by watchnerd » Sat Oct 06, 2018 2:00 am

MotoTrojan wrote:
Sat Oct 06, 2018 1:49 am
watchnerd wrote:
Fri Oct 05, 2018 11:00 pm
MotoTrojan wrote:
Fri Oct 05, 2018 9:49 pm
watchnerd wrote:
Fri Oct 05, 2018 8:56 pm
afan wrote:
Fri Oct 05, 2018 8:20 pm


Sure. But you can get that by increasing your stock allocation in VTI. No need to go to play with value funds.
Yeah, I don't get it, either.

If I want more risk / more return, I increase my equities and decrease my bonds.
Well I’m 100/0. Now what?
Leverage.
Why is that better than small-cap? Why wouldn’t I leverage a 90/10 or 60/40 instead of 100/0?
You can leverage at any mix of stocks/bonds you want.

It's better than SCV because it's guaranteed to instantly juice returns (and losses)....as opposed to the value premium which may or may not appear after decades.
Tax Sheltered: 35% US Stock | 35% ex-US Stock | 30% TTM || Taxable: 35% US Stock | 35% ex-US Stock | 15% TTM | 15% Munis

typical.investor
Posts: 253
Joined: Mon Jun 11, 2018 3:17 am

Re: Where is that value premium?

Post by typical.investor » Sat Oct 06, 2018 3:15 am

KJVanguard wrote:
Fri Oct 05, 2018 4:18 am

SG = 9.26%
SV = 8.93%

Well, this isn't the way I'm told it's supposed to be, so what is going on? Why do we have a 0.33% annual premium for Small Growth, an investment that would also have cost you less in taxes due to its lower dividends (and thus higher cap gains that get preferred tax treatment).

Let's consider the possibilities, or at least all that I can think of and you can tell me which one(s) you think are correct:

1. There is no value premium and that's exactly what we are seeing here.

2. There is a value premium, but one can't capture it with a Vanguard fund as they are not "valuey" enough. Certainly possible since Fama French seem to be talking about DEEP VALUE, something you won't find at Vanguard or substantially anywhere else for that matter. I guess one could pay Rick Ferri (being the lowest priced advisor) to get access to DEEP VALUE DFA funds, though that would cost money (even if modest) and that would eat into your value premium. I don't think Rick would even advise use of his firm ONLY to get into DFA funds. Obviously, Rick is much more than a mere DFA gate-keeper so he could add value for the average investor, though if you are reading this odds are you are well beyond average.

3. There is a value premium, but it just failed to show up over a span of "only" 20 years. Certainly possible, but knowing I'm going to win in the "long run" is of little comfort to us mere mortals who will all be dead in the long run. Hey, I mean the equity premium has failed to show up in Japan in the last three decades, but we all still believe in it. Wonder if investors in Japan believe in it, with many not being old enough to remember it?
Possibility 4: Measurement artifact.

Can anyone help me with this. I know the index has changed multiple times and is CRSP now. In any case, I have two questions:

1) The CRSP value/growth cut point is 50%. But even if a previously rated growth company has now has a 65% value score, it will remain in the growth index (as the threshold is 66.67%). Even if it then moves to 68%, it will still be allocated half to value half to growth (though it's clearly in the value range) for some time. This is to reduce movement of course. To what degree might this transaction reducing mechanism confound returns.

2) CRSP states "It is also possible for the same security to be classified differently in different market capitalization–based indexes depending on the characteristics it displays in each segment. In other words, a security may be classified as value in one capitalization-based index and as growth in another capitalization-based index, depending on its AR score within each market capitalization segment." How does this impact small companies? Is there anything systematic that say pushes what are small value companies in the total market index into growth in the small market cap index (or vice versa)?

See http://www.crsp.com/files/Equity-Indexe ... uide_0.pdf

I can't recall 100% the previous indexes, but thought they were market cap split meaning a growth boom would push what were really growth companies into the value index. For example, 10 years of MSCI had "The objective of the MSCI Value and Growth Indexes design is to divide constituents of an underlying market capitalization index into a value index and a growth index, each targeting 50% of the free float-adjusted market capitalization of the underlying index. The market capitalization of each constituent should be fully represented in the combination of the value index and the growth index, and, at the same time, should not be “double-counted”. One security may, however, be represented in both the value index and the growth index at a partial weight." http://www.crsp.com/files/Equity-Indexe ... uide_0.pdf

I mean since Jun 2000, Vanguard Small Cap Growth Index I VSGIX has had a positive loading on value at 0.05%. Ok not significant, but how can the growth and value indexes both have a positive value loading? Is it that the more lottery type growth stocks are weeded out of the index by methodology?

As far as Japan, I am not sure about your assessment as I am not holding a Japanese market portfolio, but rather value which has done better there. It's the largest segment in my small international value fund which has done reasonable well recently. It seems value rather than small explains it.

1986.1 -2011.9 https://www.mof.go.jp/english/pri/publi ... pr022c.pdf
Monthly average return by holding the market portfolio is 0.17%, which is not significantly different from zero. Excess return against risk- free rate is -0.07%. Investors are not rewarded by taking the market risk. SMB factor shows 0.03% return, and the size effect is not found in the Japanese market. On the other hand, HML factor shows 0.67% return with statistical significance. The high return by HML factor suggests the existence of risks that reward in the Japanese market. However, if investors consider the market return as a return from the stock market, then they may find very weak evidence to justify investment in the equity market.
Since Sep 2013, FNDC which is now 37% Japan has a CAGR of 8.79% vs SCHC with 19% in Japan and a CAGR of 6.90%. Both are small developed. FNDC has a value loading.

User avatar
JoMoney
Posts: 5917
Joined: Tue Jul 23, 2013 5:31 am

Re: Where is that value premium?

Post by JoMoney » Sat Oct 06, 2018 3:52 am

typical.investor wrote:
Sat Oct 06, 2018 3:15 am
...
I mean since Jun 2000, Vanguard Small Cap Growth Index I VSGIX has had a positive loading on value at 0.05%. Ok not significant, but how can the growth and value indexes both have a positive value loading? Is it that the more lottery type growth stocks are weeded out of the index by methodology?...
Stocks that aren't really investable, because of liquidity and other concerns are likely weeded out. There is a research paper that suggests that is the case, and that it benefits growth portfolios.
Portfolio Constituency Rules and the Value Premium in the Small-Cap Space

The 'Total Stock Market' by definition should have zero factor loading, but a regression on a Total Stock Market mutual fund reveals a tilt to value
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

stlutz
Posts: 4752
Joined: Fri Jan 02, 2009 1:08 am

Re: Where is that value premium?

Post by stlutz » Sat Oct 06, 2018 3:52 am

typical.investor wrote:
Sat Oct 06, 2018 3:15 am

Possibility 4: Measurement artifact.
...
I mean since Jun 2000, Vanguard Small Cap Growth Index I VSGIX has had a positive loading on value at 0.05%. Ok not significant, but how can the growth and value indexes both have a positive value loading? Is it that the more lottery type growth stocks are weeded out of the index by methodology?
It's not really a CRSP-specific thing--it's just that nature of building institutional portfolios. The "growth penalty" lies almost entirely in stocks that are considered institutionally uninvestable.

This paper (https://papers.ssrn.com/sol3/papers.cfm ... id=2394711) looked at the "base screen" of stocks that a firm like DFA would use and when that got applied to the Fama/French universe of stocks, the large value premium/growth penalty in the smallcap space shrunk dramatically.
We find that restrictive constituency rules that omit the smallest, most illiquid stocks improve the performance of both value and growth stock portfolios. However, we find the impact of constituency rule restrictions on portfolio returns to be asymmetric with respect to value and growth in the small-cap investment space. Growth portfolios benefit from these changes more than value portfolios. Consistent with prior research, we find that value and growth style portfolios constructed from more liquid equities to be void of a statistically significant value-minus-growth return premium.
That is why a "growth" index fund loads on value. Vanguard's isn't the only one to do so. Look at IJT for example.

A lot of expectations have gotten built up around what is really just a database-usage quirk.

typical.investor
Posts: 253
Joined: Mon Jun 11, 2018 3:17 am

Re: Where is that value premium?

Post by typical.investor » Sat Oct 06, 2018 5:01 am

stlutz wrote:
Sat Oct 06, 2018 3:52 am
typical.investor wrote:
Sat Oct 06, 2018 3:15 am

Possibility 4: Measurement artifact.
...
I mean since Jun 2000, Vanguard Small Cap Growth Index I VSGIX has had a positive loading on value at 0.05%. Ok not significant, but how can the growth and value indexes both have a positive value loading? Is it that the more lottery type growth stocks are weeded out of the index by methodology?
It's not really a CRSP-specific thing--it's just that nature of building institutional portfolios. The "growth penalty" lies almost entirely in stocks that are considered institutionally uninvestable.

This paper (https://papers.ssrn.com/sol3/papers.cfm ... id=2394711) looked at the "base screen" of stocks that a firm like DFA would use and when that got applied to the Fama/French universe of stocks, the large value premium/growth penalty in the smallcap space shrunk dramatically.
We find that restrictive constituency rules that omit the smallest, most illiquid stocks improve the performance of both value and growth stock portfolios. However, we find the impact of constituency rule restrictions on portfolio returns to be asymmetric with respect to value and growth in the small-cap investment space. Growth portfolios benefit from these changes more than value portfolios. Consistent with prior research, we find that value and growth style portfolios constructed from more liquid equities to be void of a statistically significant value-minus-growth return premium.
That is why a "growth" index fund loads on value. Vanguard's isn't the only one to do so. Look at IJT for example.

A lot of expectations have gotten built up around what is really just a database-usage quirk.
I see. So the value premium definitely exists and it is to be found in virtually every ETF and mutual fund because they exclude stocks considered uninvestible.

Ok, then mostly I am out the higher cost of my value funds and perhaps surrender some liquidity in the sense that value can underperform for long stretches and you generally don’t want to sell assets when down.

I guess I can live with that potential downside if it never appears. Oh, higher taxes too I guess but the methodology of the funds I use doesn’t focus on yield (and is between VB and VBR)

Nate79
Posts: 3503
Joined: Thu Aug 11, 2016 6:24 pm
Location: Portland, OR

Re: Where is that value premium?

Post by Nate79 » Sat Oct 06, 2018 5:13 am

So explain again why there is so much emphasis put to tilting to small cap value instead of small cap growth?

HEDGEFUNDIE
Posts: 876
Joined: Sun Oct 22, 2017 2:06 pm

Re: Where is that value premium?

Post by HEDGEFUNDIE » Sat Oct 06, 2018 5:16 am

Nate79 wrote:
Sat Oct 06, 2018 5:13 am
So explain again why there is so much emphasis put to tilting to small cap value instead of small cap growth?
The vale premium shows up more clearly with the S&P 600 index, IJS vs. IJT:

https://www.portfoliovisualizer.com/fun ... F05%2F2018

Although the “Cumulative Active Return” chart toward the bottom shows that the cumulative small value premium reached its high water mark in January 2007, and has been shrinking ever since.
Last edited by HEDGEFUNDIE on Sat Oct 06, 2018 5:37 am, edited 3 times in total.

typical.investor
Posts: 253
Joined: Mon Jun 11, 2018 3:17 am

Re: Where is that value premium?

Post by typical.investor » Sat Oct 06, 2018 5:19 am

Nate79 wrote:
Sat Oct 06, 2018 5:13 am
So explain again why there is so much emphasis put to tilting to small cap value instead of small cap growth?
Because hope springs eternal and any other view is detrimental to the condition of the human spirit?

User avatar
JoMoney
Posts: 5917
Joined: Tue Jul 23, 2013 5:31 am

Re: Where is that value premium?

Post by JoMoney » Sat Oct 06, 2018 5:44 am

HEDGEFUNDIE wrote:
Sat Oct 06, 2018 5:16 am
Nate79 wrote:
Sat Oct 06, 2018 5:13 am
So explain again why there is so much emphasis put to tilting to small cap value instead of small cap growth?
The vale premium shows up more clearly with the S&P 600 index, IJS vs. IJT:

https://www.portfoliovisualizer.com/fun ... F05%2F2018

Although the “Cumulative Active Return” chart toward the bottom shows that the cumulative small value premium reached its high water mark in January 2007, and has been shrinking ever since.
It wasn't doing so hot in the period leading up to the start date of those ETFs either...

Image
To repeat the Bogle quote again..
John Bogle in 1996 annual report wrote: https://www.sec.gov/Archives/edgar/data ... 000339.txt

... On this note, you may recall that I mentioned to you one year ago that the various market segments seem to enjoy cycles of superiority in their (unpredictable) turns. But, over longer periods, their respective returns tend to converge. Surely, both the interim and longer-run patterns are demonstrated in the charts above....
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

HEDGEFUNDIE
Posts: 876
Joined: Sun Oct 22, 2017 2:06 pm

Re: Where is that value premium?

Post by HEDGEFUNDIE » Sat Oct 06, 2018 5:46 am

JoMoney wrote:
Sat Oct 06, 2018 5:44 am
HEDGEFUNDIE wrote:
Sat Oct 06, 2018 5:16 am
Nate79 wrote:
Sat Oct 06, 2018 5:13 am
So explain again why there is so much emphasis put to tilting to small cap value instead of small cap growth?
The vale premium shows up more clearly with the S&P 600 index, IJS vs. IJT:

https://www.portfoliovisualizer.com/fun ... F05%2F2018

Although the “Cumulative Active Return” chart toward the bottom shows that the cumulative small value premium reached its high water mark in January 2007, and has been shrinking ever since.
It wasn't doing so hot in the period leading up to the start date of those ETFs either...

Image
Which is why I invest in IJR. Small cap premium is still going strong!

User avatar
vineviz
Posts: 1745
Joined: Tue May 15, 2018 1:55 pm

Re: Where is that value premium?

Post by vineviz » Sat Oct 06, 2018 8:10 am

HEDGEFUNDIE wrote:
Sat Oct 06, 2018 5:16 am
Nate79 wrote:
Sat Oct 06, 2018 5:13 am
So explain again why there is so much emphasis put to tilting to small cap value instead of small cap growth?
The vale premium shows up more clearly with the S&P 600 index, IJS vs. IJT:

https://www.portfoliovisualizer.com/fun ... F05%2F2018

Although the “Cumulative Active Return” chart toward the bottom shows that the cumulative small value premium reached its high water mark in January 2007, and has been shrinking ever since.
Although “value” and “growth” are widely viewed as opposites by investors (and are used as opposing labels by index providers), in a risk factor model they are not opposites.

It’s somewhat more accurate in that context to think of the value premium as being “value minus all stocks” rather than “value minus growth”.

Also the premium can gradually turn positive or it can do so suddenly. I used Fama-French data to extend the S&P 600 Value index back in time, and this chart plots the active returns versus a broad group of small cap “blend” funds. During the period shown the premium was positive but was mostly realized over the course of less than three years. If you got scared in 2000 or 2001, you probably missed most of it.

Image

In fact, given what we know about investor behavior I’d be willing to wager that even though the index had a positive premium the typical investor might very well have missed out on it.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

jclear
Posts: 106
Joined: Mon Aug 20, 2018 12:47 am

Re: Where is that value premium?

Post by jclear » Sat Oct 06, 2018 5:35 pm

afan wrote:
Fri Oct 05, 2018 8:20 pm
vineviz wrote:
Fri Oct 05, 2018 11:44 am


If my risk tolerance is higher than yours, for instance, it's completely rational for me to take on more risk than you even if (especially if) we are both getting the same risk-adjusted return.
Sure. But you can get that by increasing your stock allocation in VTI. No need to go to play with value funds.
For the same reason your entire equity portfolio isn't the S&P500. Because you don't expect that mixing that with bonds will produce the best risk-adjusted return. Maybe you have international too, for some reason.
I'm not a big fan of international investing compared to a lot here. But even I'm willing to have 10% of my equities in emerging markets, because I expect that to deliver better risk-adjusted return even if EM has the same expected return as USA TSM. For the same reason I have a REIT slice, because I expect that to produce better risk-adjusted return even if REITs have less expected return than USA TSM.

KJVanguard
Posts: 61
Joined: Fri Oct 05, 2018 3:39 am

Re: Where is that value premium?

Post by KJVanguard » Sat Oct 06, 2018 5:40 pm

Nate79 wrote:
Sat Oct 06, 2018 5:13 am
So explain again why there is so much emphasis put to tilting to small cap value instead of small cap growth?
I can easily imagine someone reasonable owning a blend fund.

I must admit that I'm curious who owns growth funds and why?

Elysium
Posts: 1261
Joined: Mon Apr 02, 2007 6:22 pm

Re: Where is that value premium?

Post by Elysium » Sat Oct 06, 2018 5:55 pm

KJVanguard wrote:
Sat Oct 06, 2018 5:40 pm
Nate79 wrote:
Sat Oct 06, 2018 5:13 am
So explain again why there is so much emphasis put to tilting to small cap value instead of small cap growth?
I can easily imagine someone reasonable owning a blend fund.

I must admit that I'm curious who owns growth funds and why?
Hmm.. may be someone who owns a value fund and a growth fund instead of a blend fund.

User avatar
watchnerd
Posts: 1428
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Where is that value premium?

Post by watchnerd » Sat Oct 06, 2018 6:01 pm

Elysium wrote:
Sat Oct 06, 2018 5:55 pm
KJVanguard wrote:
Sat Oct 06, 2018 5:40 pm
Nate79 wrote:
Sat Oct 06, 2018 5:13 am
So explain again why there is so much emphasis put to tilting to small cap value instead of small cap growth?
I can easily imagine someone reasonable owning a blend fund.

I must admit that I'm curious who owns growth funds and why?
Hmm.. may be someone who owns a value fund and a growth fund instead of a blend fund.
I was once in a 401k "orientation" class and the guy leading it, some shill from Fidelity, had people raise their hands and ask them if they had selected, or what would they select, between a blend, value, and growth fund.

About 70% raised their hands on "growth" because they wanted "it to grow."
Tax Sheltered: 35% US Stock | 35% ex-US Stock | 30% TTM || Taxable: 35% US Stock | 35% ex-US Stock | 15% TTM | 15% Munis

User avatar
Taylor Larimore
Advisory Board
Posts: 27445
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

"Don't look for the needle in the haystack. Just buy the haystack"

Post by Taylor Larimore » Sat Oct 06, 2018 6:15 pm

KJVanguard wrote:
Fri Oct 05, 2018 4:18 am
Thank you for reading. Please share your thoughts with me.
Don't look for the needle in the haystack. Just buy the haystack! -- John Bogle
Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

jalbert
Posts: 3709
Joined: Fri Apr 10, 2015 12:29 am

Re: Where is that value premium?

Post by jalbert » Sat Oct 06, 2018 6:27 pm

zonto wrote:
Fri Oct 05, 2018 4:41 pm
Thought I'd share this article from a couple years ago: Is the Value Premium Disappearing?. It also analyzes the value premium in developed and emerging international markets. Also this thread: viewtopic.php?t=191001, which analyzes the relationship between the value premium and inflation rates.
Re: Ben Carlson’s blog post cited above that discusses the int’l value premium, I don’t think you can evaluate a value premium by looking at aggregated EM it DM portfolios. There are too many other variables. For instance, the value component of EM (or DM) May have a very different country makeup than the broad index or growth component, and the different country exposure will also drive differing returns. This needs to be evaluated on an individual country basis.
Risk is not a guarantor of return.

typical.investor
Posts: 253
Joined: Mon Jun 11, 2018 3:17 am

Re: Where is that value premium?

Post by typical.investor » Sat Oct 06, 2018 7:54 pm

jalbert wrote:
Sat Oct 06, 2018 6:27 pm
zonto wrote:
Fri Oct 05, 2018 4:41 pm
Thought I'd share this article from a couple years ago: Is the Value Premium Disappearing?. It also analyzes the value premium in developed and emerging international markets. Also this thread: viewtopic.php?t=191001, which analyzes the relationship between the value premium and inflation rates.
Re: Ben Carlson’s blog post cited above that discusses the int’l value premium, I don’t think you can evaluate a value premium by looking at aggregated EM it DM portfolios. There are too many other variables. For instance, the value component of EM (or DM) May have a very different country makeup than the broad index or growth component, and the different country exposure will also drive differing returns. This needs to be evaluated on an individual country basis.
It's a good point but taken together, don't both the MSCI and French-Fama data include a pretty similar country list? Granted Korea and/or a few others may show in EM in one and developed in the other, but isn't the fact that value outperformed in both evidence that it exists.

By the way, I do 100% agree that we need to look at individual countries (or perhaps regions) for factor regressions to see accurate loadings, and that looking at aggregate data doesn't necessarily explain which fund has the most value or size or whatever exposure.

In any case, the main problem is that French/Fama isn't investible because I just don't see people trying to short EM stocks, and if they did - the costs would have to be factored in which they aren't for French/Fama data. You can point to DFA returns, but they use various screens. So I think you'd need to compare actual funds. Of course as jalbert points out, who knows how country exposure for that particular fund affects things unless you look into it.

What I kind of don't understand is why value would do well in overseas developed markets. Doesn't value typically do well in moderate and rising inflation. With the low inflation in many place, it would seem growth should be favored but I guess it's more complicated than that.

jalbert
Posts: 3709
Joined: Fri Apr 10, 2015 12:29 am

Re: Where is that value premium?

Post by jalbert » Sat Oct 06, 2018 10:53 pm

It's a good point but taken together, don't both the MSCI and French-Fama data include a pretty similar country list?
That is not the point. Comparing the return of a value portfolio aggregated over, say all of the MSCI developed market countries to the return of an MSCI developed markets total market index fund to look for excess return has the pitfall that the value portfolio may have different country weights than the total market portfolio despite the fact that both use the MSCI universe.

Consider looking for a size premium for developed markets. Small and micro cap DM stocks generally have a much larger weighting to Japanese and Canadian equities than a total market developed markets index. This makes it hard to know if you are seeing a size premium, appreciation of the yen, high commodity prices driving Canadian outperformance etc.

You have to look at single countries to evaluate factor premia properly, imho.

By the way, here is a study of factor premia in the UK:

https://papers.ssrn.com/sol3/papers.cfm ... id=3020947

The results are not encouraging.
Risk is not a guarantor of return.

Park
Posts: 554
Joined: Sat Nov 06, 2010 4:56 pm

Re: Where is that value premium?

Post by Park » Mon Oct 08, 2018 4:30 pm

About the value premium, my concern is will it persist? One explanation for the value premium is that it is due to risk, and because of that, it will persist. Another explanation is that it is due to human behavior; human behavior will persist, and there are limits to arbitraging behavioral mistakes.

I'm still concerned about whether it will persist. I'm no more than an average investor. If I know about it, then it is very common knowledge. If enough people tilt to value, then it becomes a crowded trade regardless of risk or limits to arbitrage.

I've heard other arguments about why I shouldn't be worried. I'm writing this post so others who are more knowledgeable can correct me as indicated on the following. If value is becoming a crowded trade, there should be an increase in value returns, and that hasn't happened. Also, if value is becoming a crowded trade, then the dispersion between value and growth stocks should have decreased, and that hasn't happened.

User avatar
watchnerd
Posts: 1428
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Where is that value premium?

Post by watchnerd » Mon Oct 08, 2018 7:06 pm

Will the 'small' part persist given the massive decline in the number of publicly traded companies, most of them small?

As Bogle points out, we used to have the Wilshire 5000 index....no longer. Less than 4,000 publicly traded companies now.
Tax Sheltered: 35% US Stock | 35% ex-US Stock | 30% TTM || Taxable: 35% US Stock | 35% ex-US Stock | 15% TTM | 15% Munis

iamlucky13
Posts: 1010
Joined: Sat Mar 04, 2017 5:28 pm
Location: Western Washington

Re: Where is that value premium?

Post by iamlucky13 » Mon Oct 08, 2018 8:07 pm

watchnerd wrote:
Mon Oct 08, 2018 7:06 pm
Will the 'small' part persist given the massive decline in the number of publicly traded companies, most of them small?

As Bogle points out, we used to have the Wilshire 5000 index....no longer. Less than 4,000 publicly traded companies now.
It's rather hard to even guess since as a starting point we don't know if the decline will continue or reverse.

From about 10 minutes searching, I didn't find data further back than 1975, so I don't even know if the count today is atypically low, normal, or perhaps even atypically high historically.

However, with twice as many new international stocks being listed as US stocks delisted in the time since the US peak, I have a suspiscion about globalization of trade being a factor in the number of US stocks.

http://www.nber.org/digest/sep15/w21181.html
Image

User avatar
watchnerd
Posts: 1428
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Where is that value premium?

Post by watchnerd » Mon Oct 08, 2018 8:58 pm

iamlucky13 wrote:
Mon Oct 08, 2018 8:07 pm
watchnerd wrote:
Mon Oct 08, 2018 7:06 pm
Will the 'small' part persist given the massive decline in the number of publicly traded companies, most of them small?

As Bogle points out, we used to have the Wilshire 5000 index....no longer. Less than 4,000 publicly traded companies now.
It's rather hard to even guess since as a starting point we don't know if the decline will continue or reverse.

From about 10 minutes searching, I didn't find data further back than 1975, so I don't even know if the count today is atypically low, normal, or perhaps even atypically high historically.

However, with twice as many new international stocks being listed as US stocks delisted in the time since the US peak, I have a suspiscion about globalization of trade being a factor in the number of US stocks.

http://www.nber.org/digest/sep15/w21181.html
Image
THAT is interesting and data I've never seen before.....
Tax Sheltered: 35% US Stock | 35% ex-US Stock | 30% TTM || Taxable: 35% US Stock | 35% ex-US Stock | 15% TTM | 15% Munis

iamlucky13
Posts: 1010
Joined: Sat Mar 04, 2017 5:28 pm
Location: Western Washington

Re: Where is that value premium?

Post by iamlucky13 » Tue Oct 09, 2018 8:40 am

watchnerd wrote:
Mon Oct 08, 2018 8:58 pm
THAT is interesting and data I've never seen before.....
I actually hadn't either, but when you raised your point, I had a suspicion I might see a graph like that if anybody had published such data. I found that article and graph almost instantly, and quite a few others that were similar. Most of them, unfortunately, focused on roughly 1990 to present, I suppose because they wanted to highlight the decline.

KJVanguard
Posts: 61
Joined: Fri Oct 05, 2018 3:39 am

Re: Where is that value premium?

Post by KJVanguard » Tue Oct 09, 2018 4:40 pm

Hello Taylor. I've been a Vanguard investor since January 1995 when I invested the minimum in Total Stock Market Index. In the decades since TSM has become my largest holding, so I do indeed own the whole haystack.

My 2nd largest US equity holding would be Vanguard Small Value Index, which was bought with the proceeds from Tax-Managed Small-Cap (which I inherited & subsequently sold for a tax loss).

I am pretty much forced to keep TSM due to my holding being almost 50% unrealized gains. I own SV Index because even if I am wrong and it never outperforms, I find it hard to imagine that 852 stocks can lag the market by much when they are a meaningful portion of the market. And I have yet to see any research that shows that small & value underperforms.

I wish there were a way to get a greater Small & Value tilt though there is no cost effective alternative IMO.

User avatar
Taylor Larimore
Advisory Board
Posts: 27445
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Re: Where is that value premium?

Post by Taylor Larimore » Tue Oct 09, 2018 7:27 pm

KJVanguard wrote:
Tue Oct 09, 2018 4:40 pm
I am pretty much forced to keep TSM due to my holding being almost 50% unrealized gains.
KJVanguard:

"almost 50% unrealized gains" tells me that TSM has been a winning fund for you.

Stay the course and keep investing simple.

Read my "Simplicity" link below.

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

pascalwager
Posts: 1216
Joined: Mon Oct 31, 2011 8:36 pm

Re: Where is that value premium?

Post by pascalwager » Wed Oct 10, 2018 1:04 am

KJVanguard wrote:
Sat Oct 06, 2018 5:40 pm
Nate79 wrote:
Sat Oct 06, 2018 5:13 am
So explain again why there is so much emphasis put to tilting to small cap value instead of small cap growth?
I can easily imagine someone reasonable owning a blend fund.

I must admit that I'm curious who owns growth funds and why?
I don't know who owns them, but John Cochrane says the proper owners are investors who wish to reduce risk.

Post Reply